Research Title |
A GEOGRAPHIC INFORMATION SYSTEM (GIS)-BASED ANALYSIS |
Date of Distribution |
22 October 2013 |
Conference |
Title of the Conference |
the 34th Asian Conference on Remote Sensing 2013 |
Organiser |
Indonesian Remote Sensing Society and Asian Association on Remote Sensing |
Conference Place |
Discovery Kartika Plaza HotelIndonesian |
Province/State |
Bali – Indonesia |
Conference Date |
20 October 2013 |
To |
24 October 2013 |
Proceeding Paper |
Volume |
34 |
Issue |
- |
Page |
4250-4257 |
Editors/edition/publisher |
|
Abstract |
The aim of this study was to create a model to predict a geographic area suitable for rubber
tree cultivation with Extreme Learning Machine, an effective learning scheme of feedforward Neural
Networks, and Decision Tree, under GIS-based analysis in the Udon Thani province. The study was
composed of three main steps. The first step was to determine biophysical characteristics of land for
rubber planting, as modified from the principle of the FAO 1983 guideline for land evaluation. The
second step was to create a model to predict suitable land by using the results from the first step. The
final step was to determine the validity of the results obtained from the model by comparing and
measuring the efficiency and precision by using the K-fold cross validation. The prediction by
Extreme Learning Machine and Decision Tree revealed accuracies of 99.490%, and 99.780%,
respectively. The data implied that Decision Tree showed a better accuracy than Extreme Learning
Machine. |
Author |
|
Peer Review Status |
ไม่มีผู้ประเมินอิสระ |
Level of Conference |
นานาชาติ |
Type of Proceeding |
Full paper |
Type of Presentation |
Poster |
Part of thesis |
true |
Presentation awarding |
false |
Attach file |
|
Citation |
0
|
|